21:17:34 <kevinbenton> #startmeeting networking
21:17:35 <openstack> Meeting started Mon Jun 26 21:17:34 2017 UTC and is due to finish in 60 minutes.  The chair is kevinbenton. Information about MeetBot at http://wiki.debian.org/MeetBot.
21:17:36 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
21:17:38 <openstack> The meeting name has been set to 'networking'
21:18:15 <kevinbenton> #topic announcements
21:18:23 <kevinbenton> #link https://releases.openstack.org/pike/schedule.html
21:18:55 <kevinbenton> this week is the pike-3 milesonte
21:18:58 <kevinbenton> milestone*
21:19:14 <mlavalle> isn't it next month?
21:19:25 <kevinbenton> arg!
21:19:28 <kevinbenton> yes :)
21:19:36 <kevinbenton> good thing
21:19:38 <kevinbenton> i panicked
21:19:53 <mlavalle> it is a good thing indeed.... yaay!
21:20:00 <kevinbenton> this week is keystone feature proposal freeze then
21:20:05 <kevinbenton> :)
21:20:27 <kevinbenton> #topic Bugs and gate failures
21:21:12 <kevinbenton> looks like bug deputy was jlibosva
21:21:21 <kevinbenton> mlavalle: are you aware of any big bugs?
21:21:52 <mlavalle> not that I am aware of
21:22:23 <mlavalle> not new anyway
21:23:21 <kevinbenton> ok
21:23:42 <kevinbenton> and it looks like amotoki_away volunteered for bug deputy this week
21:23:49 <mlavalle> correct
21:23:57 <kevinbenton> ok
21:23:58 <mlavalle> during last week's meeting
21:24:09 <kevinbenton> i know red hat is having some special event this week
21:24:27 <kevinbenton> and armax is on vacation
21:24:36 <mlavalle> ahhh, that is why we are missing Ihar
21:24:42 <kevinbenton> so we'll just go to open discussion to talk about whatever
21:24:46 <kevinbenton> #topic open discussion
21:25:15 <mlavalle> I just want to mentions that we have started the move of the docs to the Neutron tree
21:25:31 <mlavalle> Yesterday I pushed a patchset to move the Networking Guide
21:26:03 <mlavalle> and this morning I had the very pleasent surprise that someone else had pushed a similar patch one day eralier
21:26:07 <mlavalle> earlier
21:26:35 <mlavalle> It is a good thing, because it seems we have one new net contributor commited to help with the move
21:26:37 <kevinbenton> mlavalle: cool. is it ready to go or are we waiting on stuff from the docs side?
21:26:59 <mlavalle> he is still working through someissues
21:27:19 <mlavalle> he has 2 patchsets, one for the networking guide and the other for the admin bits
21:27:36 <mlavalle> I'll keep my eye on them
21:27:45 <mlavalle> and will help as needed
21:28:21 <mlavalle> hang on, let me post the links to the reviews
21:28:47 <mlavalle> #link https://review.openstack.org/#/c/477498/
21:29:28 <kevinbenton> excellent
21:29:34 <kevinbenton> the target for this is by Pike, right?
21:29:35 <mlavalle> #link https://review.openstack.org/#/c/477105/
21:29:53 <mlavalle> yes, that is why it is important to keep a close eye on this
21:30:07 <mlavalle> amotoki_away is also being very helpful reviewing
21:30:56 <mlavalle> that's all I have in this regard
21:31:05 <slaweq> hello
21:31:32 <slaweq> may I ask about one thing if there will be no any other topics?
21:32:17 <kevinbenton> slaweq: go ahead
21:32:33 <slaweq> so I wanted to ask about tempest test for qos rule types
21:32:49 <slaweq> I found that one test is in fact broken and is not checking anything currently
21:33:24 <slaweq> so I made simple patch to fix that but YAMAMOTO pointed that it might break other deployments tests, see https://review.openstack.org/#/c/477192/1/neutron/tests/tempest/api/test_qos.py
21:33:29 <kevinbenton> slaweq: yeah, i saw that
21:33:50 <slaweq> so my question is: what You think about testing just returned keys in such test?
21:34:06 <slaweq> because values are in fact dependent of deployment
21:34:07 <kevinbenton> slaweq: so it's not dependent on which drivers are loaded, right?
21:34:28 <slaweq> kevinbenton: values are dependent on drivers
21:35:03 <slaweq> and similar issue I have now in another patch https://review.openstack.org/#/c/475260/ also
21:35:11 <kevinbenton> slaweq: i mean what you just asked to change to is one not dependent right?
21:35:30 <kevinbenton> because yamamoto is right, we should uncouple this from specific gate deployment topology as much as possible
21:35:51 <slaweq> yes, if we would just test in tempest if there are proper keys returned by neutron server than it would be not dependent on loaded drivers
21:36:20 <slaweq> but if we are checking what values are returned (available rule_types) then it depends on neutron server config
21:36:30 <kevinbenton> slaweq: right
21:36:33 <kevinbenton> slaweq: that's what i meant before
21:36:48 <kevinbenton> i think that's what we should do for now
21:37:11 <kevinbenton> a fullstack test might make more sense for testing specific drivers
21:37:18 <slaweq> IMHO we can test if values are proper in fullstack tests where we can specify directly loaded drivers
21:37:21 <slaweq> :)
21:37:43 <slaweq> and there is already such test for rule_types in fullstack AFAIR
21:38:20 <kevinbenton> ok cool
21:38:34 <kevinbenton> slaweq: so go ahead and update that patch to just check keys
21:38:40 <slaweq> kevinbenton: ok, so I will change this tempest test to check only keys
21:38:44 <slaweq> thx for help
21:38:54 <kevinbenton> ok, anything else?
21:39:02 <mlavalle> not from me
21:39:25 <slaweq> I'm done :)
21:40:12 <kevinbenton> ok
21:40:14 <kevinbenton> thanks!
21:40:18 <kevinbenton> #endmeeting